Scientists from (where else?) UC Berkeley and the University of Oxford in Geneva, working under a National Science Foundation grant, have been able to extract PIN codes and banking information directly from the brains of human subjects by using an over-the-counter computer interface.
According to the scientists’ research paper, submitted to the 21st USENIX Security Symposium, an electroencepholography-based brain-computer interface (or BCI) such as those produced by Emotiv Systems and NeuroSky, costing around $300, can be used to extract sensitive user information from unsuspecting subjects.
The BCIs are designed to allow hands-free computing by letting a user control his machine with thoughts. These commercial applications are the developmental descendants of aircraft control systems the U.S. military was developing at least as early as the 1990s.
A few years ago, there was even a simple toy called a Force Trainer, produced by Uncle Milton Industries, which let a child use his mind to turn on and off a small fan that would float a plastic ball inside a tube.
Similar devices have been the subject of science fiction for many years, including the Clint Eastwood movie “Firefox,” which featured a thought-controlled supersonic aircraft.
In the Berkeley experiments, the scientists sat subjects in front of computer screens that flashed images of faces, banks and numbers. By recording the response from the BCIs, particularly a channel of signal called P300, the scientists were able to increase the possibility of correctly guessing the subjects’ private information.
The P300 signal was focused on because it is usually produced when a person recognizes someone or something significant that they interact with all the time.
The method wasn’t totally accurate, but correct deductions on the first guess were high enough that a brain hacker could definitely find the effort worthwhile. For example, a PIN could be guessed accurately the first time in 20 percent of cases.
Possibly more insidious, however, is the potential use the researchers found for the method in lie detection and interrogation. The method could tell interrogators whether certain information existed in a person’s brain.
The paper found that this use of the method was susceptible to fewer countermeasures than traditional lie-detection tests.
It doesn’t take a lot of imagination to come up with ways for this technology to be misused by Big Brother.